RERTR 2012 ― 34th INTERNATIONAL MEETING ON

REDUCED ENRICHMENT FOR RESEARCH AND TEST REACTORS

October 14-17, 2012

Warsaw Marriott Hotel

Warsaw, Poland

OVERVIEW OF LANL PROGRESS IN PROCESS DEVELOPMENT,

ADVANCED CHARACTERIZATION METHODS AND PROTOTYPE

FABRICATION

David E. Dombrowski

Materials Science and Technology Division, Metallurgy (MST-6)

Los Alamos National Laboratory, PO Box 1663, Los Alamos, NM 87544 - USA

ABSTRACT

Recent progress at LANL in process development, advanced characterization method

development, prototype fabrication and alternate process development for monolithic LEU fuel

will be discussed. Process development studies included HIP (Hot Isostatic Pressing) can design

development and plasma spraying of zirconium. Progress on design and testing of formed HIP

cans will be presented. Advanced characterization method development concentrated on

interface bond strength measurement. Fracture energy results from both bulge testing and

notched minicantilever testing will be presented. Maps of zirconium thickness on cold rolled

LEU foils were performed using x-ray fluorescence. Residual stress measurements were made

using synchrotron x-ray radiation and the incremental slitting method. The processing which

produced two prototypical LEU foils will be discussed.

Introduction

This paper provides an overview of work at LANL in process development, development of

bond strength measurement, residual stress measurements and prototype development over the

last year. Process development was oriented towards producing full scale prototypes, resolving

scale up issues, and developing process variations such as canless HIP which have potential for

fuel production cost savings. The process development overview will include casting, rolling,

HIP can fabrication, canless HIP, plasma spraying, laser ablation and promotion of aluminum

grain growth across the HIP (Hot Isostatic Pressing) bond line. The bond strength measurement

method overview will include the controlled bulge test and mini-cantilever test. The residual

stress overview will include recent synchrotron measurements, and incremental slitting (crack

compliance method) measurements. Selected topics will be discussed in more depth in separate

posters and papers.

Casting Development for RERTR-FE Foil Fabrication

The initial casting done at LANL was done with the minimization of LEU machining in mind.

The original plan was to cast a large ingot, homogenize the ingot, and bare hot roll the ingot to a

thickness which was suitable for machining into separate co-rolling coupons. A number of

issues with the shape of the hot rolled ingot and subsequent machining prompted interest in

casting closer to the final coupon thickness. A combination of heat flow and fluid flow modeling

(Flow3D computational fluid dynamics) and in-situ monitoring of casting and mold temperatures

with more than ten thermocouples lead to an improved three part mold design for three zone

vacuum induction melting (VIM) and casting process. Table 1 shows the mold geometry

iterations used and the high (> 90%) U-10Mo metal yield achieved for the optimized mold

geometry. Figure 1 shows an ingot from the optimized mold with thickness measurements. This

ingot was cast to a very thin and uniform dimension (5 mm).

Table 1 Iterations in mold geometry and resulting

Height Width Thickness Height Width Thickness

(mm) (mm) (mm) (mm) (mm) (mm)

DU-7Mo 203 203 13 51 227 18 9.606 9.006 0.938

DU-10Mo 203 203 13 51 227 18 9.502 8.960 0.943

DU-10Mo 206 155 17 99 188 23 13.918 12.965 0.932

DU-10Mo 206 155 17 99 188 23 13.900 13.275 0.955

LEU-10Mo 108 148 14 83 148 18 4.395 3.625 0.825

DU-10Mo 105 140 14 72 140 18 4.399 4.256 0.967

DU-10Mo 115 295 5 80 295 10 4.404 3.878 0.881

DU-10Mo 115 295 5 80 295 10 4.379 4.150 0.948

DU-10Mo 115 295 5 80 295 10 4.401 4.101 0.932

DU-10Mo 115 295 5 80 295 10 4.411 4.099 0.929

DU-10Mo 115 295 5 80 295 10

DU-10Mo 115 295 5 80 295 10

Yield

fraction

Nominal

Compositi

Mold Dimensions Hot Top Dimensions

Charge

Wt (kg)

Casting

Wt (kg)

RERTR-FE Foil Fabrication

An LEU U-10Mo casting (11K-635) was

bare hot rolled using a salt bath (650 C),

flattened at room temperature, homogenized

in vacuum (1000C for four hours), machined

into individual co-rolling coupons. Two

Figure 1 Photograph of DU-10Mo Ingot

12K-641 and schematic showing ingot

thickness measurements. Pour temperature

1328C. Mold top was 1200C. Bottom of

mold cavity was 750C. R. Aikin

machined coupons were canned, co-rolled (650C), de-canned and cold rolled using INL rolling

procedure [1]. The co-rolled LEU foils are shown in Figure 2a after decanning. Foil 11K-635-A

had a split or defect which was sheared off prior to cold rolling. Figures 2b and 2c show the two

foils after cold rolling. Foil 11K-635-A was cold rolled to a fuel meat thickness of 0.33 mm, and

foil 11K-635-B1 was cold rolled to a fuel meat thickness of 0.5 mm. No edge defects or

cracking was observed. It should be noted that cold rolled foil 11K-635-A was stored in a plastic

bag in a laboratory with naturally low humidity for over eight months and showed no indications

of cracking or bowing.

a.

c.

Figure 2 Photographs of LEU U-Mo foils. a. Co-rolled foils after decanning. b. Cold rolled

foil 11k-635-B1. Overall dimensions are 0.737 m x 81 mm x 0.67 mm. c. Cold rolled foil 11k-

635-A. Overall dimensions are 1.26 m x 74 mm x 0.38 mm. D. Hammon, K. Clarke, J. Scott

Figure 3 shows the thickness of the roll bonded zirconium (Zr) layer on one surface of foil 11K-

635-A as measured using a calibrated, handheld x-ray fluorescence unit. Zr thickness varied

between 0.020 and 0.022 mm. Some of the mapping shows low Zr thickness at the edges, but

this is an artifact attributed to the overlap of the foil and tabletop detector in the detector field of

view. Greater accuracy resolution of the Zr thickness at the edges will be obtained by use of the

macro-XRF unit.

Figure 3 Thickness of the roll bonded Zr layer on one surface of foil 11K-635-A. D. Summa

HIP Can Development

Several initiatives have been started to reduce the cost of HIP can fabrication, including use of

formed HIP cans and canless HIP. The present method of HIP can fabrication in the program is

to have a trained welder build each six sided HIP can by hand from six sheets of stainless steel.

This is a labor intensive process and the weld volume and location often leads to significant

dimensional distortions even before the can is loaded with foils and cladding. One method of

minimizing the weld volume and labor time is to use a can which uses a stainless steel sheet

mechanically formed into a five sides of a HIP can. The contents could be loaded from the open

b.

top, and only the top would have to be welded on to complete the HIP can. An example of a

formed can is shown in Figure 4. This example and results are described in a separate poster and

paper at the conference [2].

a. b.

Figure 4 Example of a formed HIP can. a. Round formed HIP cans used to demonstrate

feasibility b. Schematic of hydroforming press used to make round formed HIP cans. K. Clarke

Canless HIP is an attempt to decrease HIP can fabrication costs and bonding force distribution

issues by eliminating the stainless steel HIP can altogether. The concept is to seal the U-10Mo

foil in aluminum cladding under vacuum before it is placed in the HIP. Electron beam welding

is used to weld the perimeter of the top and bottom aluminum cladding together. Since the

electron beam welding process is performed in vacuum, there is a vacuum between the U-10Mo

foil and the cladding after the aluminum cladding pieces have been welded together. A

schematic of the process is shown in Figure 5a. Electron beam welding trials were made on 610

mm long 6061 aluminum cladding and surrogate foil (stainless steel) assemblies. Figure 6b

shows two plates after e-beam welding and HIP. The plates show a small amount of thickness

decrease and no inflation. This indicates that there was no major leak at the electron beam

welds. Metallographic examination will be used to confirm aluminum to aluminum bonding and

aluminum to stainless steel bonding. More details are available in a separate poster and paper at

the conference [2]

a. b.

Figure 5 Canless HIP a. Process schematic b. Two plate assemblies (Al/stainless steel/ Al)

after electron beam welding and HIP. C. Cross

Grain growth across the Al-Al bond line in HIP’d monolithic fuel plates is desirable but has yet

to be attained; at least 25% grain growth across the bond line is desired. To improve this, studies

of the effect of aluminum cladding surface preparation on the bond lines of HIP’d Al-Al

sandwiches are being performed. Electron back scatter diffraction (EBSD) is being used to

characterize the aluminum grain growth. A comparison of two different cleaning processes on

the bond line is shown in Figure 6. More details are available in a separate poster and paper at

the conference [2]

a. b.

Figure 6 EBSD characterization of bond line for two different surface cleaning processes. a.

LANL cleaning produced 0.4% grain growth at bond line (none visible here) b. Babcock

&Wilcox cleaning process produced 1.8% grain growth seen in circled region. R. Hackenberg

Plasma Spraying a Diffusion Barrier Coating on U-Mo Foils

The standard way of applying a zirconium diffusion barrier coating on U-Mo foils is a roll

bonding method called co-rolling. Applying these coatings by plasma spraying may decrease U-

10Mo recycling costs because all the rolling scrap would be uncoated U-Mo. Uncoated U-Mo is

less expensive to recycle than zirconium coated U-Mo. Use of plasma spraying would also

remove the canning costs associated with co-rolling. Low pressure plasma spraying (LPPS) has

been used to produce zirconium coatings and molybdenum coatings on stainless steel foils and

U-10Mo foils. Uniformity of a zirconium coating on DU-10Mo foil is shown by optical

metallography in Figure 7. Spot measurements of zirconium thickness were made using a

handheld x-ray fluorescence (XRF) unit. These are shown in Figure 8. Details on surface

roughness, grain size and morphology, and scale up to a 610 mm long foil are available in a

separate poster and paper at the conference [3].

Figure 7 Optical metallography image of plasma sprayed zirconium thickness and porosity. K.

Hollis

bond

line

Figure 8 Spot measurements of plasma sprayed zirconium thickness on two different foil

materials. K. Hollis, D. Summa

Integrated Intelligent Machining

Machining of monolithic fuel plates to final thickness can be difficult and costly if there are

deviations in foil position or shape, or aluminum thickness. In the worst case, these factors could

combine and result machining of the aluminum below the minimum aluminum cladding

thickness (Al MinClad) which is required for safe reactor operation. Integrated Intelligent

Machining combines x-ray fluorescence, pulsed echo, time-of-flight and ultrasonic phased array

measurement methods to provide real time aluminum thickness (Al MinClad) guidance during

plate machining. For aluminum thicknesses above 0.8 mm, ultrasonic testing provides fast,

accurate results. An ultrasonic probe such as the ultrasonic squirter system shown in Figure 9

can be directly attached to the machining head and used for real time measurements on

aluminum thickness. Such a system could with appropriate calibration use normal machining

coolants. For aluminum thicknesses below 0.8 mm, a portable x-ray fluorescence source as

shown in Figure 9 can provide more accurate aluminum thickness readings because the

ultrasonic methods have more scatter in their measurements due to surface roughness.

Calibration of XRF measurements is done using attenuation curves as shown in Figure 10.

Aluminum composition affects calibration. For very thin aluminum (< .125 mm), fluorescence

absorption is dominated by intervening Zr layer. For aluminum thickness > .65 - .75 mm,

internal scattering becomes more pronounced

Figure 9 Integrated Intelligent Machining combines x-ray fluorescence, pulsed echo, time-of-

flight and ultrasonic phased array measurement methods for real time MinClad guidance during

plate machining. D. Summa.

Bond Strength Characterization

Bond strength measurements and ideally fracture

toughness values are needed for fuel qualification as

well as for development of quality control standards.

Bond strength measurements for thin materials such

as fuel plates are very challenging because there is

not enough material to grip and apply mechanical

force. There are three bond strengths needed: the

aluminum to aluminum bond, the aluminum to

zirconium bond, and the U-Mo to zirconium bond.

Two methods have been under development at

LANL, the controlled bulge test and the

minicantilever test. The controlled bulge test uses a

fluid under an interface to cause a deflection in the plate and ultimately a delamination at the

desired interface. That round upward deflection is quantitatively characterized by using Digital

Image Correlation (DIC). An energy analysis yields bond strength and interfacial fracture.

Figure 11 shows a schematic cross section of the fixture used for measurements, the initial data

obtained, and the conceptual breakdown of total energy into components including the

delamination energy. More details are available in a separate poster and paper at the conference

[4].

Figure 11 Controlled bulge test a. Schematic cross section of plate in testing fixture and

pressure versus volume curve obtained from test b. Schematic description of energy analysis

used to obtain the delamination energy. C. Liu, M. Lovato

Figure 10 Al thickness calibration requires

examination of ratio of detected Zr to detected U

signals. D. Summa

Figure 12 The load displacement curve for a minicantilver test is shown on the left. The inset

shows a DIC strain map. a. The mini-cantilever test for an Al/Al interface begins with deflection

of the cantilever with a nanoindenter. b. Final deflection or fracture. N. Mara, W. Mook

Previous work [5] had developed a micro-cantilever which was able to measure interface

strength of all three interfaces. However that technique used a Focused Ion Beam to make the

microcantilevers and nanoindenter to test the beams. It would not be feasible in present hot cells.

The mini-cantilever test was designed to utilize much simpler equipment to machine the

cantilever and test it. The method has been used for monolithic Al minicantilevers made from

HIP’d Al, minicantilevers machined from an Al/Al interface and minicantilevers machined from

Al/Zr interfaces. Although the minicantilevers can be tested in small load frames in a hot cell,

initial work was done in-situ in an SEM for diagnostic purposes. Figure 12 shows a typical test

for an Al/Al interface minicantilever. More details are available in a separate poster and paper at

the conference [4].

Residual Stress Measurement

Residual stress measurements have been performed using synchrotron radiation and the

incremental slitting (compliance) method. Synchotron measurements were made of residual

stress in INL miniplates. Each miniplate had a different combination of foil annealing (no

annealing or annealed) and cooling rate at the end of a HIP cycle (slow cooling or fast cooling).

Results are shown in Figure 13. In-plane strains are generally similar for the different combinations

of process conditions with one exception. The high gradient of in-plane strain near the plate end

which is observed for the fast cooling sample is not observed for slow cooling samples. More details

are available in a separate poster and paper at the conference [4].

Residual stress measurements have been made as a function of depth using the incremental slitting

method on two HIP’d plates containing DU-Mo foils. The plates are each nominally 610 mm long

and 76 mm wide. The thin plate was nominally 1.6 mm thick and the thick plate was nominally 3

mm thick. The experimental set up is shown in Figure 14. A preliminary plot of residual stress

versus depth has been generated. Final results will be available once microscopy confirms the actual

thicknesses of Al, Zr and DU-Mo at the slit location for each plate. Preliminary results indicate that

there is a pattern of compressive stress in the DU-Mo. The residual stress magnitudes appear to be

limited by low strength of 6061 Al after thermal processing in the HIP. More details are available

in a separate poster and paper at the conference [4].

Al-Al interface

Figure 13 Residual stress maps calculated from synchrotron x-ray radiation of miniplates.

Figure 14 Experimental set up for performing incremental slitting measurements using electric-

discharge machining (EDM). A pair of transducers on the side opposite the slitting is used for

primary data acquisition.

Acknowledgements

The authors would like to acknowledge the financial support of the US Department of Energy

Global Threat Reduction Initiative Reactor Convert program. Los Alamos National Laboratory, an

affirmative action equal opportunity employer, is operated by Los Alamos National Security, LLC,

for the National Nuclear Security Administration of the U.S. Department of Energy under contract

DE-AC52-06NA25396.

References

1. G. A. Moore, M. C. Marshall, Co-Rolled U10Mo/Zirconium-Barrier-Layer Monolithic Fuel

Foil Fabrication Process, INLreport INL/EXT-10-17774, January 2010

2. Kester D. Clarke, Carl E. Cross, Robert E. Hackenberg, Rodney J. McCabe, Joel D.

Montalvo, Matthew J. Dvornak, Randall L. Edwards, Justin M. Crapps, Ralph R. Trujillo,

Beverly Aikin, Victor D. Vargas, Kendall J. Hollis, Bogdan Mihaila, Duncan Hammon,

Richard W. Hudson, Tim J. Tucker, Jeffrey E. Scott, Andrew N. Duffield, Richard Y.

Weinberg, David E. Dombrowski, Development of Aluminum-Clad Fuel Plate Processing

Through Canned and Canless Hot Isostatic Pressing (HIP), and Studies of Aluminum

Cladding Grain Growth during HIP, RERTR 2012, October 14-17 2012, Warsaw, Poland

3. K.J. Hollis, Plasma Spraying of Diffusion Barrier Coatings for LEU Monolithic Fuel,

RERTR 2012, October 14-17 2012, Warsaw, Poland

4. C. Liu, M.L. Lovato, D.J. Alexander, K.D. Clarke, N.A. Mara, W.M. Mook, M.B. Prime,

D.W. Brown, Experimental Investigation of Bonding Strength and Residual Stresses in LEU

Fuel Plate, RERTR 2012, October 14-17 2012, Warsaw, Poland

5. N. Mara, J. Crapps, T. Wynn, K. Clarke, P. Dickerson, B. Mihaila, D. Dombrowski,

Nanomechanical Behavior of U-10Mo/Zr/Al Fuel Assemblies, October 23-27, 2011, RERTR

2011, Santiago, Chile